Problem Set 4: Solutions

UNIVERSITY OF ALABAMA Department of Physics and Astronomy PH 126 / LeClair Fall 2009 Problem Set 4: Solutions 1. Two resistors are connected in parallel, with values R 1 and R 2. A total current I o divides somehow between them. Show that the condition I 1 + I 2 I o, together with the requirement of minimum power dissipation, leads to the same current values that we would calculate with normal circuit formulas. This illustrates a general variational principle that holds for direct current networks: the distribution of currents within the networks, for a given input current I o, is always that which gives the least total power dissipation. First, let s figure out the current in each resistance using the normal circuit formulas. Since the two resistors are in parallel, they will have the same potential difference across them, but in general different currents (unless R 1 R 2, in which case the currents are the same). Let I 1 and I 2 be the currents in resistors R 1 and R 2, respectively, with the total current then given by conservation of charge, I o I 1 + I 2. Given the current through each resistor, we can readily calculate the voltage drop on each, which again must be the same for both resistors: V 1 I 1 R 1 V 2 I 2 R 2 V 1 V 2 I 1 R 1 I 2 R 2 We can find the current in each resistor from the known total current I o by noting that I o I 1 + I 2, and thus I 2 I o I 1 I 1 R 1 I 2 R 2 I 1 R 1 (I o I 1 ) R 2 I 1 R 1 + I 1 R 2 I o R 2 I 1 R 2 R 1 + R 2 I o [ 1 1 + R 1 R 2 ] I o Thus, the fraction of the total current in first resistor depends on the ratio of the two resistors. The larger resistor 2 is, the more current that will flow through the first resistor not shocking! Given the above expression for I 1, we can easily find I 2 from I 2 I o I 1, which yields I 2 R 1 R 1 + R 2 I o [ 1 1 + R 2 R 1 ] I o

Our derivation of the currents in each resistor has so far only relied on conservation of energy (components in parallel have the same voltage) and conservation of charge (I o I 1 + I 2 ), we have not invoked any special laws about combining parallel resistors. In fact, that is what we have just derived! Now for the requirement of minimum power dissipation. We want to find the distribution of currents that results in minimum power dissipation in the most general way, specifically not using the results of the previous portion of this problem. We will only assume that resistors R 1 and R 2 carry currents I 1 and I 2, respectively, and that these two currents add up to the total current, I o I 1 + I 2. In other words, we only assume conservation of charge to start with. The total power dissipated is just the sum of the individual power dissipations in the two resistors: P tot P 1 + P 2 I 2 1R 1 + I 2 2R 2 I 2 1R 1 + (I o I 1 ) 2 R 2 I 2 1 (R 1 + R 2 ) + I 2 or 2 2IR 2 I 1 For the last part, we invoked our conservation of charge condition (I o I 1 + I 2 ). What to do next? We have now the total power P tot in both resistors as a function of the current in R 1. If we minimize the total power with respect to I 1, we will have found the value of I 1 which leads to the minimum power dissipation. Since I 2 is then fixed by the total current I once we know I 1, I 2 I o I 1, this is sufficient to establish the values of both I 1 and I 2 that lead to minimum power dissipation. Of course, to find the minimum of P tot for any value of I 1, we need to take a derivative i... dp tot di 1 2I 1 (R 1 + R 2 ) 2I o R 2 0 I 1 R 2 R 1 + R 2 I o Lo and behold, the minimum power dissipation occurs when the currents are distributed exactly as we expect for parallel resistors. At this point, you can easily find I 2 as well, given I 2 I o I 1. The general rule is that current in a dc circuit distributes itself such that the total power dissipation is minimum, which we will not prove here. Of course... hold on a minute. We missed one small point: by finding dp tot di 1 and setting it to zero, we have certainly found an extreme value for P tot. We did not prove whether it is a maximum or a minimum however. This is exactly the sort of shenanigans we need to avoid to do proper physics, so we should apply the second derivative test. i Keep in mind that the total current I o is fixed, so di o /di 1 0. And, yes we should technically be using partial derivatives here (differentiating with respect to I 1 while holding everything else constant), but since only I 1 varies that would be a bit pedantic. Plus, the symbols seem to scare people.

Recall briefly that after finding the extreme point of a function f(x) via df/dx xa 0, one should calculate d 2 f/dx 2 xa : if d 2 f/dx 2 xa < 0, you have a maximum, if d 2 f/dx 2 xa > 0 you have a minimum, and if d 2 f/dx 2 xa 0, the test basically wasted your time. Anyway: d 2 P tot di 2 1 2 (R 1 + R 2 ) > 0 Since resistances are always positive, we have in fact found a minimum of P tot. Crisis averted. Don t let that lull you into complacency, however: you always need to apply the second derivative test to see what you ve really found. At the very least, you should invoke the symmetry of the function to justify having found a minimum or maximum, and not just take derivatives and set them to zero all willy-nilly. 2. Show that if a battery of fixed internal voltage V and internal resistance r is connected to a variable external resistance R the maximum power is delivered to the external resistor when rr. The circuit we are considering is just a series combination of the (ideal) internal voltage source V, the internal resistance R i, and the external resistance R. Since all the elements are in series, the current is the same in each, which we will call I. Applying conservation of energy, V IR IR i 0 I V R + R i The power delivered to the external resistor P R is just I 2 R: ( V P R I 2 R R + R i ) R ( V) 2 R (R + R i ) 2 Similar to the last problem, we can maximize the power delivered to the resistor R by differentiating the power with respect to R and setting the result equal to zero: dp R dr d [ ( V) 2 dr 1 (R + R i ) 2 2R (R + R i ) 3 1 2R R + R i R + R i 2R R i R ] [ R (R + R i ) 2 ( V) 2 1 (R + R i ) 2 + 2R (R + R i ) 3 ] 0 The power is indeed extremal when the external resistor matches the internal resistance of the battery.

Again, we apply the second derivative test to see whether this is a maximum or a minimum. First, let s find the second derivative, and simplify it as much as possible. d 2 P R dr 2 d ( )] [( V) 2 1 dr (R + R i ) 2 2R (R + R i ) 3 [ ] ( V) 2 2 (R + R i ) 3 2 (R + R i ) 3 + 6R (R + R i ) 4 [ ] ( V) 2 4 (R + R i ) 3 + 6R (R + R i ) 4 [ ] ( V)2 6R (R + R i ) 3 4 R + R i We are concerned with the value of the second derivative at the point RR i, the extreme point: d 2 P R dr 2 RRi [ ] ( V)2 6Ri (R i + R i ) 3 4 ( V)2 R i + R i 8R 3 [3 4] ( V)2 i 8R 3 i < 0 The second derivative is always negative, so we have found a maximum. Thus, the power delivered to an external resistor is maximum when RR i. 3. A resistor R is to be connected across the terminals A, B of the circuit below. (a) For what value of R will the power dissipated in the resistor be the greatest? To answer this, construct the Thévenin equivalent circuit and then invoke the result of the previous problem. (b) How much power will be dissipated in R? 15Ω 120V - + 10Ω B A R 10Ω 4. Two graphite rods are of equal length. One is a cylinder of radius a. The other is conical, tapering linearly from a radius a at one end to radius b at the other. Show that the end-to-end electrical resistance of the conical rod is a/b times that of the cylindrical rod. Hint: consider the rod to be made up of thin, disk-like slices, all in series.

The cylindrical conductor is trivial: if it is of radius a and length l, and has resistivity ρ, then R cyl ρl πa 2 (1) Of course, we don t know the length l or resistivity ρ, but they will not matter in the end. What about the cone? Break the cone up into many disks of thickness dx. Stacking these disks up with increasing radius can build us a cone: a b ( ) b a r(x)a + l x 2r x0 xl dx If we start out with a radius a at one end of the cone, and the other end has a radius b, then the radius as a function of position along the cone is easily determined. Let our origin (x 0) be the end of the cone with radius a, and assume the cone has a total length l, same as the cylinder. Again, we will not need this length in the end, but it is convenient now. The radius at any position along the cone is then ( ) b a r(x) a + x l (2) If the current is in the x direction, then each infinitesimally thick disk is basically just a tiny segment of wire with thickness dx and cross-sectional area π [r(x)] 2. If we assume the same resistivity ρ, the resistance of each disk must be dr cone ρdx π [r(x)] 2 ρdx π ( a + ( b a l ) x ) 2 (3) The total resistance of the cone is found by integrating over all such disks, from x 0 to the end of the cone at xl. For convenience, let c(b a) /l. l R cone dr cyl 0 ρdx π (a + cx) 2 ρ [ π 1 c (a + cx) ρ [ 1 0 π cb 1 ] ρ [ ] b a ca πc ab ] l ρl πab (4)

Here we have a nice result: the resistance of a cone is the same as a resistance of a cylinder whose radius is the geometric mean cone s radii. That is, if we substitute a 2 ab in our usual formula for the resistance of a cylinder, we have the result for a cone. Anyway: the desired result now follows readily, R cone R cyl a b (5) 5. A laminated conductor was made by depositing, alternately, layers of silver 10 nm thick and layers of tin 20 nm thick. The composite material, considered on a larger scale, may be considered a homogeneous but anisotropic material with electrical conductivity σ for currents perpendicular to the planes of the layers, and a different conductivity σ for currents parallel to that plane. Given that the conductivity of silver is 7.2 times that of tin, find the ratio σ /σ. First, let us sketch out the situation given: t2 t1 a b see UCSD07 solns, quicker way with numbers Now, let s solve the situation in a more general way. We are not told how many layers of each type we have, and it will not matter in the end. For now, however, assume we have n 1 layers of tin of conductivity σ 1 and n 2 layers of silver of conductivity σ 2. Instead of conductivity, we can equivalently use resistivity ρ when it is more convenient, with ρ 1/σ. We will also say the tin layers have thickness t 1, and the silver layers thickness t 2. The total thickness of our entire multi-layer stack is then t tot n 1 t 1 + n 2 t 2. First, consider the perpendicular conductivity, the case where we pass current upward through the stack, perpendicular to the planes of the layers. When a current is flowing, electrons pass through each layer in sequence, and we can consider the stack of layers to be resistors in series. If the layers have an area of A ab (see the Figure above) and a thickness t 1 or t 2, we can readily calculate the resistance presented by a single tin or silver layer with current perpendicular to the layers:

R 1, ρ 1t 1 A t 1 σ 1 A R 2, ρ 2t 2 A t 2 σ 2 A For reasons that should become apparent below, it will be convenient in this case to work with the resistivity rather than the conductivity, and invert the result later. The total resistance of the stack is then just a series combination of n 1 resistors of value R 1 and n 2 resistors of value R 2 : R tot, n 1 R 1, + n 2 R 2, 1 A (ρ 1t 1 n 1 + ρ 2 t 2 n 2 ) If we measure the whole stack and find this resistance, we can define an effective resistivity or conductivity for the whole stack in terms of the total resistance and total thickness of the multilayer. If the resistivity of the whole stack for perpendicular currents is ρ 1/σ, then: R tot, ρ t tot A ρ AR tot, t tot Now we just need to plug in what we know and simplify... ρ AR tot, t tot A [ ] 1 t tot A (ρ 1t 1 n 1 + ρ 2 t 2 n 2 ) ρ ρ 1t 1 n 1 + ρ 2 t 2 n 2 t tot ρ 1t 1 n 1 + ρ 2 t 2 n 2 n 1 t 1 + n 2 t 2 We can simplify this somewhat if we realize that we have the same number of silver and tin layers - we are told that the layers are deposited alternatingly. If we let n 1 n 2 n bi, meaning we count the number of bilayers instead, then t tot n bi (t 1 + t 2 ), and ρ n biρ 1 t 1 + n bi ρ 2 t 2 n bi t 1 + n bi t 2 ρ 1t 1 + ρ 2 t 2 t 1 + t 2 This is a nice, simple result: for current perpendicular to the planes, the effective resistivity is just a thickness-weighted average of the resistivities of the individual layers. Given the resistivity in the perpendicular case, we can now find the conductivity σ σ 1 t 1 + t 2 t 1 + t 2 ρ ρ 1 t 1 + ρ 2 t t 1 2 σ 1 + t σ 1σ 2 (t 1 + t 2 ) 2 σ σ 2 t 1 + σ 1 t 2 2 As a consistency check, we can take a couple of limiting cases. First, let σ 1 σ 2 σ. This corresponds to a homogeneous lump of a single material, and we find σ σ, as expected. Next, we can check for

σ 1 0. In this case, one layer is not conducting at all, and since the layers are in series, this means no current flows through the stack at all, and σ 0 as expected. Finally, we notice that the number of bilayers is irrelevant. Since the layers do not affect each other in our simple model of conduction, there is no reason to expect otherwise. So far so good. What other limiting cases can you check? Next, let us consider current flowing parallel to the plane of the layers, from (for example) left to right in the figure above. Now the stack looks like many parallel resistors. A single tin layer of thickness t 1 and in-plane dimensions a and b now presents a resistance R 1, ρ 1a t 1 b a t 1 bσ 1 Similarly, each silver layer presents a resistance R 2, ρ 2a t 2 b a t 2 bσ 2 One bilayer of silver and tin means a parallel combination of these two resistances: 1 R bi, 1 R 1, + 1 R 2, b a (t 1σ 1 + t 2 σ 2 ) If we have n bi bilayers, then the total equivalent resistance is easily found: 1 R tot, n bi 1 R bi, n bi b a (t 1σ 1 + t 2 σ 2 ) Given the total resistance, we can now calculate the conductivity directly (in this case, first finding the resistivity does not save us any algebra), noting that the length of the whole stack along the direction of the current is just a, and the cross-sectional area is bt tot bn bi (t 1 + t 2 ): σ 1 ρ a n bi (t 1 + t 2 ) br tot, a n bi (t 1 + t 2 ) b ( ) b n bi (t 1 σ 1 + t 2 σ 2 ) t 1σ 1 + t 2 σ 2 a t 1 + t 2 Again, a sensible result: the effective conductivity for current parallel to the planes is just a thicknessweighted average of the conductivities of the individual layers. Again, you can convince yourself with a couple of limiting cases that this result makes some sense. Now that we have both parallel and perpendicular conductivities, we can easily find the anisotropy σ /σ.

σ /σ ρ σ 1σ 2 (t 1 + t 2 ) t 1 + t 2 ρ σ 2 t 1 + σ 1 t 2 σ 1 t 1 + σ 2 t 2 σ 1 σ 2 (t 1 + t 2 ) 2 (t 1 σ 1 + t 2 σ 2 ) (t 1 σ 2 + t 2 σ 1 ) Finally, we are given that the conductivity of silver is 7.2 times that of tin, and the tin layers thickness is twice that of the silver. Thus, t 1 2t 2 and σ 2 7.2σ 1. The actual values and units do not matter, as this is a dimensionless ratio (you should verify this fact... ), and you should find σ /σ 0.457. And, once again, you can check that for σ 1 σ 2, we have σ /σ 1, as it must if both materials are the same. 6. The circuit at right is known as a Wheatstone Bridge, and it is a useful circuit for measuring small changes in resistance. Perhaps you can figure out why. Three of the four branches on our bridge have identical resistance R, but the fourth has a slightly different resistance, R more than the other branches, such that its total resistance is R + R. In terms of the source voltage V s, base resistance R and change in resistance R, what is the potential difference between points a and b? You may assume the voltage source and wires are perfect (no internal resistance and no voltage drop, respectively). R R Vs a b R R+ΔR Wheatstone Bridge Label the nodes on the bridge a-d, as shown in the figure below, and let a current I 1 flow from point d through a to c, and a current I 2 flow from d through b to c. Looking more carefully at the bridge, we notice that it is nothing more than two sets of series resistors, connected in parallel with each other. This immediately means that the voltage drop across the left side of the bridge, following nodes d a c, must be the same as the voltage drop across the right side of the bridge, following nodes d b c. Both are V dc, and both must be the same as the source voltage: I R I1 d I2 R Vs a b R c R+δR Labeling notes and currents in the Wheatstone Bridge

V dc V s. If we can find the current in each resistor, then with the known source potential difference we will know the voltage at any point in the circuit we like, and finding V ab is no problem. Let the current from the source V s be I. This current I leaving the source will at node a split into the separate currents I 1 and I 2 ; conservation of charge requires II 1 + I 2. At node c, the currents recombine into I. On the leftmost branch of the bridge, the current I 1 creates a voltage drop I 1 R across each resistor. Similarly, on the rightmost branch of the bridge, the resistor R has a voltage drop I 2 R and the lower resistor has a voltage drop I 2 (R + δr). Equating the total voltage drop on each branch of the bridge: V s I 1 R + I 1 R I 2 R + I 2 (R + δr) I 1 V s 2R V s I 2 2R + δr Now that we know the currents in terms of known quantities, we can find V ab by walking" from point a to point b and summing the changes in potential difference. Starting at node a, we move toward node d against the current I 1, which means we gain a potential difference I 1 R. Moving from node d to node b, we move with the current I 2, which means we lose a potential difference I 2 R. Thus, the total potential difference between points a and b must be ( Vs V ab I 1 R I 2 R R (I 1 + I 2 ) R 2R ) ( V s V ab V s ( 1 2 R R + δr δr 4R + 2δR V s 2R + δr ) ) If the change in resistance δr is small compared to R (δr R), the term in the denominator can be approximated 4R + δr 4R, and we have V ab 1 4 V s ( ) δr R (δr R) Thus, for small changes in resistance, the voltage measured across the bridge is directly proportional to the change in resistance, which is the basic utility of this circuit: it allows one to measure small changes on top of a large base resistance. Fundamentally, it is a difference measurement, meaning that one directly measures changes in the quantity of interest, rather than measuring the whole thing and trying to uncover subtle changes. This behavior is very useful for, e.g., strain gauges, temperature sensors, and many other devices. 7. A dead battery is charged by connecting it to the live battery of another car with jumper cables (see

below). Determine the current in the starter and in the dead battery. 0.01! 1! + + - 12 V 10 V - 0.06! starter live battery dead battery Since this circuit has several branches and multiple batteries, we cannot reduce it by using our rules of series and parallel resistors - we have to use our general circuit rules (Kirchhoff s rules). In order to do that, we first need to assign currents in each branch of the circuit. It doesn t matter what directions we choose at all, assigning directions is just to define what is, relatively speaking, positive and negative. If we choose the direction for one current incorrectly, we will get a negative number for that current to let us know. Below, we choose initial currents I 1, I 2, and I 3 in each branch of the circuit. I1 I2 I3 R1 R2 + - + - V1 R3 V2 Here we have also labeled each component symbolically to make the algebra a bit easier to sort out we don t want to just plug in numbers right away, or we ll make a mess of things. Note that since we have three unknowns the three currents so we will need three equations to solve this problem completely. We have three possible loops (left, right, outer), which gives us two equations, and two junctions, which gives us one more. If we have N loops or N junctions, we get N 1 equations from either. Now we are ready to apply the rules. First, conservation of charge (the junction rule ). We have only two junctions in this circuit, in the center at the top and bottom where three wires meet. The junction rule basically states that the current into a junction (or node) must equal the current out. In the case of

the upper node, this means: I 1 I 2 + I 3 (6) You can easily verify that the lower node gives you the same equation. Next, we can apply conservation of energy (the loop rule ). There are three possible loops we can take: the rightmost one containing R 3 and R 2, the leftmost one containing R 1 and R 2, and the outer perimeter (containing R 1 and R 3 ). We only need to work through two of them - we have already one equation above, and we only need two more. Somewhat arbitrarily, we will pick the right and left side loops. First, the left-hand side loop. Start just above the live battery V 1, and walk clockwise around the loop. We cross the battery from negative to positive for a gain in potential energy, and we cross R 1 and R 3 in the direction of current flow for a loss of potential energy. These three have to sum to zero for a closed loop: I 1 R 1 I 2 R 2 + V 2 + V 1 0 (7) Next, the right-hand side loop. Again, start just above the battery (V 2 this time), and walk clockwise around the loop. Now we cross R 2 against the current and R 3 with the current for a gain and loss of voltage, respectively, and then cross the battery in the wrong direction for a potential drop: I 2 R 2 I 3 R 3 V 2 0 (8) Now we have three equations and three unknowns, and we are left with the pesky problem of solving them for the three currents. There are many ways to do this, we will illustrate two of them. Before we get started, let us repeat the three questions in a more consistent form. I 1 I 2 I 3 0 R 1 I 1 + R 2 I 2 V 1 + V 2 R 2 I 2 R 3 I 3 V 2 The first way we can proceed is by substituting the first equation into the second: V 1 + V 2 R 1 I 1 + R 2 I 2 R 1 (I 2 + I 3 ) + R 2 I 2 (R 1 + R 2 ) I 2 + R 1 I 3 Now our three equations look like this:

I 1 I 2 I 3 0 (R 1 + R 2 ) I 2 + R 1 I 3 V 1 + V 2 R 2 I 2 R 3 I 3 V 2 The last two equations now contain only I 1 and I 2, so we can solve the third equation for I 2 : I 2 I 3R 3 + V 2 R 2... and plug it in to the second one: ( ) I3 R 3 + V 2 V 1 + V 2 (R 1 + R 2 ) I 2 + R 1 I 3 (R 1 + R 2 ) + R 1 I 3 R 2 (V 1 + V 2 ) R 3 (R 1 + R 2 ) I 3 + V 2 (R 1 + R 2 ) + R 1 R 2 I 3 V 1 R 2 V 2 R 1 (R 1 R 3 + R 2 R 3 + R 1 R 2 ) I 3 I 3 R 2 V 1 R 1 V 2 R 1 R 2 + R 2 R 3 + R 1 R 3 169 A There is a sort of pleasing symmetry to the analytical answer. Now that you know I 3, you can plug it in the expression for I 2 above, you should find I 2 20 A, and similarly you can find I 1 189 A R 2 Optional: There is one more way to solve this set of equations using matrices and Cramer s rule, ii if you are familiar with this technique. If you are not familiar with matrices, you can skip to the next problem - you are not required or necessarily expected to know how to do this. First, write the three equations in matrix form: R 1 R 2 0 0 R 2 R 3 1 1 1 I 1 I 2 I 3 ai V V 1 + V 2 V 2 0 The matrix a times the column vector I gives the column vector V, with the matrices defined thusly: R 1 R 2 0 a 0 R 2 R 3 I 1 1 1 I 1 I 2 I 3 V ii See Cramer s rule in the Wikipedia to see how this works. V 1 + V 2 V 2 0

Now we can use the determinant of the matrix a with Cramer s rule to find the currents. For each current, we construct a new matrix, which is the same as the matrix a except that the column of a corresponding to that current is replaced the column vector V. Thus, for I 1, we replace column 1 in a with V, and for I 2, we replace column 2 in a with V. We find the current then by calculating the determinant of the new matrix and dividing it by det a. Below, we have highlighted the columns in a which have been replaced to make this more clear: V 1 + V 2 R 2 0 V 2 R 2 R 3 0 1 1 I 1 det a R 1 V 1 + V 2 0 0 V 2 R 3 1 0 1 I 2 det a R 1 R 2 V 1 + V 2 0 R 2 V 2 1 1 0 I 3 det a Now we need to calculate the determinant of each new matrix, and divide that by det a. iii determinant of a. First, the det a R 1 R 2 R 1 R 3 R 2 R 3 (R 1 R 2 + R 2 R 3 + R 1 R 3 ) We can now find the currents readily from the determinants of the modified matrices and det a we just found. Thankfully, the matrices have enough zeros that it is relatively easy. In case your memory is rusty, here is the determinant of an arbitrary 3 3 matrix: a 1 a 2 a 3 b 1 b 2 b 3 (a 1 b 2 c 3 a 1 b 3 c 2 ) + (a 2 b 3 c 1 a 2 b 1 c 3 ) + (a 3 b 1 c 2 a 3 b 2 c 1 ) c 1 c 2 c 3 Given this, I 1 (V 1 + V 2 ) R 2 R 3 (V 1 + V 2 ) + R 2 V 2 (R 1 R 2 + R 2 R 3 + R 1 R 3 ) (V 1 + V 2 ) R 3 + R 2 V 1 R 1 R 2 + R 2 R 3 + R 1 R 3 189 A I 2 R 1V 2 (V 1 + V 2 ) R 3 (R 1 R 2 + R 2 R 3 + R 1 R 3 ) (V 1 + V 2 ) R 3 + R 1 V 2 R 1 R 2 + R 2 R 3 + R 1 R 3 20 A I 3 R 1V 2 + R 2 V 2 (V 1 + V 2 ) R 2 (R 1 R 2 + R 2 R 3 + R 1 R 3 ) R 2 V 1 R 1 V 2 R 1 R 2 + R 2 R 3 + R 1 R 3 169 A These are the same results you would get by continuing on with the previous plug-n-chug method. Both numerically and symbolically, we can see from the above that I 1 I 2 +I 3 : iii Again, the Wikipedia entry for determinant is quite instructive.

I 2 + I 3 (V 1 + V 2 ) R 3 + R 1 V 2 + R 2 V 1 R 1 V 2 R 1 R 2 + R 2 R 3 + R 1 R 3 (V 1 + V 2 ) R 3 + R 2 V 1 R 1 R 2 + R 2 R 3 + R 1 R 3 I 1 8. A hair dryer intended for travelers operates at 115 V and also at 230 V. A switch on the dryer adjusts the dryer for the voltage in use. At each voltage, the dryer delivers 1000 W of heat. (a) What must the resistance of the heating coils be for each voltage? (b) For such a dryer, sketch a circuit consisting of two identical heating coils connected to a switch and the power outlet. Opening and closing the switch should give the proper resistance for each voltage. (c) What is the current in the heating elements at each voltage? 9. Two capacitors, one charged and the other uncharged, are connected in parallel. (a) Prove that when equilibrium is reached, each carries a fraction of the initial charge equal to the ratio of its capacitance to the sum of the two capacitances. (b) Show that the final energy is less than the initial energy, and derive a formula for the difference in terms of the initial charge and the two capacitances. This problem is easiest to start if you approach it from a conservation of energy & charge point of view. We have two capacitors. Initially, one capacitor stores a charge Q 1i, while the other is empty, Q 2i 0. After connecting them together in parallel, some charge leaves the first capacitor and goes to the second, leaving the two with charges Q 1f and Q 2f, respectively. Now, since there were no sources hooked up, and we just have the two capacitors, the total amount of charge must be the same before and after we hook them together: Q i Q f Q 1i + Q 2i Q 1f + Q 2f (given Q 2i 0) Q 1i Q 1f + Q 2f We also know that if two capacitors are connected in parallel, they will have the same voltage V across them: V f Q 1f C 1 Q 2f C 2 The fraction of the total charge left on the first capacitor can be found readily combining what we have: Q 1f Q i Q 1f Q 1i Q 1f Q 1f + Q 2f Q 1f Q 1f + C 2 C 1 Q 1f C 1 Q 1f C 1 Q 1f + C 2 Q 1f C 1 C 1 + C 2

The second capacitor must have the rest of the charge: Q 2f Q i 1 C 1 C 1 + C 2 C 2 C 1 + C 2 That was charge conservation. We can also apply energy conservation, noting that the energy of a charged capacitor is Q 2 /2C: E i E f Q 2 1i 2C 1 Q2 1f 2C 1 + Q2 2f 2C 2 The final energy can be simplified using the result of the first part of the problem - we note that Q 1f Q i C 1 / (C 1 + C 2 ) and Q 2f Q i C 2 / (C 1 + C 2 ) E f Q2 1f + Q2 2f 2C 1 2C 2 ( ) Qi C 2 ( ) 1 1 Qi C 2 2 1 + C 1 + C 2 2C 1 C 1 + C 2 2C 2 Q 2 i C 1 2 (C 1 + C 2 ) 2 + Q 2 i C 2 2 (C 1 + C 2 ) 2 Q2 i (C 1 + C 2 ) 2 (C 1 + C 2 ) 2 Q 2 i 2 (C 1 + C 2 ) ( ) ( ) Q2 i C1 C1 E i 2C 1 C 1 + C 2 C 1 + C 2 Thus, the final energy will be less than the initial energy, by a factor C 1 / (C 1 + C 2 ) < 1.